Sustainability Research Institute

Social Simulation Conference 2017

19.10.2017 - 10:40

Nick Roxburgh reports back from Social Simulation Conference 2017:

The Social Simulation Conference 2017 took place at University College Dublin at the end of September. Thanks to an Sustainability Research Institute bursary, I was able to attend the conference which represents the premier event of the social simulation calendar in Europe - an event which is attended by many of the biggest names in the field.

The conference, which takes place annually, focuses on understanding complex social problems through use of computational simulation. This broad remit meant researchers from an array of social science fields were in attendance. Alongside the session on social ecology in which I presented, there were sessions on topics such as economics, finance, democracy, regulation, innovation, society, cooperation, fisheries, violence, and demography. A number of sessions also addressed methodological issues which had relevance across the disciplinary fields.

My talk on representation of intra-household dynamics in agent-based models of smallholder communities took place during the penultimate day of the conference. I used the first part of the talk to discuss the importance of intra-household dynamics in shaping household decision-making outcomes and I spoke about how social, economic and environmental stressors may affect household members differently depending on their age, gender, and role within their household. At present, almost all agent-based models of smallholder communities use households as the lowest level of agent rather than individuals. Consequently, intra-household dynamics have rarely been manifestly modelled. My argument was that this can potentially lead to unrepresentative behaviour and ignorance about the likely unequal effects of model scenarios at an individual level.

In the second part of my talk I presented an approach that can be used to model intra-household dynamics - an approach which I have developed as part of my PhD research looking at the impact of social, economic, and environmental stressors on smallholder communities in Nepal. I then had the chance to discuss the framework I had set out with some of the leading names in my particular sub-field - an immensely valuable experience for me.

Other highlights of the conference included a talk by Wander Jager on “Exploring the variability in a social-ecological system caused by alternative formalizations of human decision-making,” a talk on model validation by Gary Polhill, a challenging and provocative keynote by Bruce Edmonds on “The post-truth drift in social simulation,” and an engaging keynote by Kathleen Carley on diffusion in networks. Another standout talk was given by the University of Leeds’ Nick Malleson on the use of data assimilation for forecasting short-term urban dynamics. The opportunity to attend the conference, present my own work, and hear talks by fellow agent-based modellers, has been an immense honour and one I am very grateful to SRI for providing me.